X-Authentication-Warning: delorie.com: mail set sender to geda-help-bounces using -f
X-Recipient: geda-help AT delorie DOT com
Message-ID: <4F6B1276.80507@laserlinc.com>
Date: Thu, 22 Mar 2012 07:52:22 -0400
From: Joshua Lansford <Joshua DOT Lansford AT laserlinc DOT com>
User-Agent: Mozilla/5.0 (X11; Linux i686; rv:10.0.2) Gecko/20120216 Thunderbird/10.0.2
MIME-Version: 1.0
To: geda-help AT delorie DOT com
Subject: Re: [geda-help] Segfault in gnetlist
References: <CAHBDEvYK2DRe7fm4Xq+5H=c7--s_eLHds3qa-K0RUNxrS7ei1Q AT mail DOT gmail DOT com>
In-Reply-To: <CAHBDEvYK2DRe7fm4Xq+5H=c7--s_eLHds3qa-K0RUNxrS7ei1Q@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Reply-To: geda-help AT delorie DOT com

Hi Tom,
   Was looking though the list and saw your request and it doesn't look 
like you have a response.
I haven't taken a detailed look at your schematic, but I have found that 
gnetlist will segfault whenever there is an empty property in the 
schematic or one of the symbols.  It made me panic the first time I hit 
it.  Now fixing the problem is routine.  This grep command like this 
should find it, and then you can remove it with vim or another text editor.
grep -n =\\s*$ *.sch
~Joshua

On 03/16/2012 08:54 PM, Thomas Oldbury wrote:
> Trying to load a schematic into xgsch2pcb results in a blank PCB file.
>
> Traced it down to a segfault here:
>
>  $ gnetlist -q -g gsch2pcb -o FreeBox.new.pcb -m gnet-gsch2pcb-tmp.scm 
> power_main.sch
>  Segmentation fault
>
> I can delete a small section of my schematic and it goes away 
> (generally around the lower optocoupler area); however, if I replace 
> it and delete another section around the same area, it also goes away! 
> Also, it seems (although it may just be me) that deleting parts 
> individually doesn't work... you have to delete all at once?! This has 
> made trying to find an individually faulty part hard, so I think it's 
> probably some combination of factors that I'm unlucky enough to trip over.
>
> I have attached my schematic in case it helps, it uses a few custom 
> symbols but the bug should still show up.
>
> Is there a way for me to track down what part is causing the 
> problem... never had this before with gEDA.
>
> Thanks,
> Tom